## Glass fragments reassembling themselves by chance

For the discussion of the sciences. Physics problems, chemistry equations, biology weirdness, it all goes here.

Moderators: gmalivuk, Moderators General, Prelates

HonoreDB
Posts: 161
Joined: Tue Feb 06, 2007 4:32 pm UTC
Contact:

### Glass fragments reassembling themselves by chance

I knock a glass off a table. It falls, and shatters into a specific arrangement of fragments, each of which contains a specific amount of heat energy in a specific configuration, and so on. While the fact that it would shatter was predictable in advance, if we describe the exact outcome specifically enough it becomes very unlikely. Call this very specific outcome "event one."

I gather up the fragments and sweep them off the same table at the same height. As they land, they reassemble themselves, purely by chance, back into an identical glass. Call this highly surprising outcome "event two."

Which event is more probable?

Dopefish
Posts: 855
Joined: Sun Sep 20, 2009 5:46 am UTC
Location: The Well of Wishes

### Re: Glass fragments reassembling themselves by chance

Surely theres infinitely many possible configurations for the (whole glass/fragments of glass) to fall into, and thus both would have probability 0, and thus would be equally probable.

Although things are kindve wierd when dealing with infinity and probability at once, since probability 0 != "could never happen", so I'm not sure if theres differing degrees of probability 0. I think in this case they'd be the same degree (if multiple degrees even do exist) though and thus they have equal probabilties of occuring, specifically, 0.

Charlie!
Posts: 2035
Joined: Sat Jan 12, 2008 8:20 pm UTC

### Re: Glass fragments reassembling themselves by chance

One, because of the requirement that the shards be swept off the table. I think that coming from a broken glass makes shards less likely than average to all move in the same direction, and so by time reversal symmetry all falling in the same direction is less likely to correspond to a glass.
Some people tell me I laugh too much. To them I say, "ha ha ha!"

BlackSails
Posts: 5315
Joined: Thu Dec 20, 2007 5:48 am UTC

### Re: Glass fragments reassembling themselves by chance

Its probably roughly the same probability, since each outcome is a single point in phase space.

big boss
Posts: 589
Joined: Tue Mar 23, 2010 7:59 am UTC

### Re: Glass fragments reassembling themselves by chance

couldn't you employ an entropy argument? Since entropy always increases and outcome one has more entropy (it is more disordered), therefore outcome 1 is more likely than the ordered state of outcome 2.
"Starbuck, what do you hear?"
"Nothing but the Rain."

iChef
Posts: 343
Joined: Fri Oct 15, 2010 9:33 pm UTC
Location: About 5 cm. south of the ring finger, USA.

### Re: Glass fragments reassembling themselves by chance

Wouldn't one be much more likely than a completely random shattering due to scratches on the glass, flaws in in from when it was made, the height at which it was dropped. When tending bar I've seen nearly identical (coming from the same factory, used and stored in the same manner) glasses break in a nearly identical pattern (around the middle with a split up the side due to how they were stacked which causes stress on the center of the glass). While it is highly unlikely that the exact same break happens it is still far, far more likely than having the glass re-assemble (which my boss would love, my glassware broker, not so much).
Those whom God loves, he must make beautiful, and a beautiful character must, in some way, suffer.
-Tailsteak author of the Webcomics 1/0 and Leftover Soup

BlackSails
Posts: 5315
Joined: Thu Dec 20, 2007 5:48 am UTC

### Re: Glass fragments reassembling themselves by chance

big boss wrote:couldn't you employ an entropy argument? Since entropy always increases and outcome one has more entropy (it is more disordered), therefore outcome 1 is more likely than the ordered state of outcome 2.

You cant use an entropy argument, since we are singling out a single microstate. Thermodynamics is for ensembles.

JWalker
Posts: 222
Joined: Sun Nov 23, 2008 1:13 am UTC

### Re: Glass fragments reassembling themselves by chance

Event one is more probable than event two. Event two is classically impossible as the shattering of the glass distorts the resulting shards such that even if you put them back together by hand you wont be able to reassemble the glass. The pieces will not fit together correctly.

Little Richie
Posts: 127
Joined: Tue Feb 05, 2008 3:02 am UTC
Location: Orlando, Florida

### Re: Glass fragments reassembling themselves by chance

JWalker wrote:Event half-two is more probable than event three. Event three is classically impossible as the shattering of the glass distorts the resulting shards such that even if you put them back together by hand you wont be able to reassemble the glass. The pieces will not fit together correctly.

In this argument wouldn't these changes you are describing be reversed in event three by the same forces that caused them in half-two? When you are reassembling the glass by hand, you are not bringing the extra energy of the fall into the system.
With man gone, will there be hope for gorilla?
Check out my blog, my mission to save the world.

Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7604
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

### Re: Glass fragments reassembling themselves by chance

JWalker wrote:Event half-two is more probable than event three. Event three is classically impossible as the shattering of the glass distorts the resulting shards such that even if you put them back together by hand you wont be able to reassemble the glass. The pieces will not fit together correctly.

Nah, it's purely an statistical effect that the local distortions tend not to match up. The odds of ending up distorted into a single particular non-fitting configuration are farily comparable to the odds of being distorted back to a fitting configuration. There are just many more non-fitting configurations than fitting ones.

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

### Re: Glass fragments reassembling themselves by chance

I still go by the statistics of observation... I've never observed a glass reassembling.
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

Glass Fractal
Posts: 497
Joined: Thu May 13, 2010 2:53 am UTC

### Re: Glass fragments reassembling themselves by chance

Technical Ben wrote:I still go by the statistics of observation... I've never observed a glass reassembling. ;(

This would be my response, too. Glass can't reassemble just from two pieces bumping into each other.

gmalivuk
GNU Terry Pratchett
Posts: 26824
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

### Re: Glass fragments reassembling themselves by chance

Technical Ben wrote:I still go by the statistics of observation... I've never observed a glass reassembling. ;(
You've never observed a glass shattering in *exactly* the same way twice, either. Thus, as already stated, both events are of comparable likelihood.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

sikyon
Posts: 344
Joined: Sun Dec 07, 2008 5:45 pm UTC

### Re: Glass fragments reassembling themselves by chance

Comparable. I would even go so far as to say that event 3 is more likely, given that a whole glass is at a more energetically favorable state than a shattered glass (less surface energy).

Bonooru1234
Posts: 12
Joined: Fri Oct 15, 2010 12:21 am UTC

### Re: Glass fragments reassembling themselves by chance

event one is more likely, eventually you will predict where the glass shards will land.
but when a glass breaks the top pieces land with the bottom pieces, this means that they will be at the same level when they fall and they will hit the ground at the same time, this means that it is impossible for the glass to reform because for it to do that would require the top to fall slower than the bottom

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

### Re: Glass fragments reassembling themselves by chance

gmalivuk wrote:
Technical Ben wrote:I still go by the statistics of observation... I've never observed a glass reassembling. ;(
You've never observed a glass shattering in *exactly* the same way twice, either. Thus, as already stated, both events are of comparable likelihood.

Huh. I'm happy with no 2 events being the same. How does that show event 3 is more likely?

In a mathematical model, where we only include atomic positions, and ignore everything from gravity to air friction, then yes, every position of atoms is as likely as the other. But a "glass" falling off a "table" make me think 100% in real world terms. In which case, there are more forces working against the glass coming together, than it falling apart.

[double edit]
I think my main confusion has been down to the cheese graters and the numbers being swapped. Sorry.
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

Seli
Posts: 17
Joined: Wed Sep 17, 2008 10:01 am UTC

### Re: Glass fragments reassembling themselves by chance

Atoms in bulk glass are distributed differently from the atoms in the surface layer, on breaking the glass the atoms at the new surfaces are re-distributed. This would have to be reversed in spontaneous reassembly, which probably makes spontaneous re-assembly orders of magnitude more unlikely than any specific breakage pathway. Entropy for the win.

Charlie!
Posts: 2035
Joined: Sat Jan 12, 2008 8:20 pm UTC

### Re: Glass fragments reassembling themselves by chance

The fact that you said "entropy" should be a red flag that the problem probably accounts for that.
Some people tell me I laugh too much. To them I say, "ha ha ha!"